Every way to check your Claude Code, Codex, Cursor, Gemini and Copilot usage limits in 2026

You're deep in a refactor, the agent is finally on a roll, and then it stops: limit reached, try again later. The annoying part isn't the limit. It's that you had no idea it was coming.

Every AI coding tool now has some way to show you where you stand. They just all hide it in different places. Here's every option I know of, from built-in commands to menu bar apps, so you can pick the one that fits how you work.

Built-in commands (free, already installed)

1. Claude Code: /usage

Type /usage (or its alias /cost) in a session. On Pro, Max, Ultra, Team and Enterprise plans it shows plan usage bars, activity stats and a breakdown of what counts against your limits, plus recommendations for cutting usage. On recent versions you can press f to let Claude Code apply the fixes it can make itself.

The Session block at the top shows token stats for the current session. If you're on a subscription, ignore the dollar figure there; it's computed at API list price. Docs

2. Codex CLI: /status and /usage

/status shows the active model, approval policy and current token usage for the session. /usage opens account usage, and /usage daily, /usage weekly or /usage cumulative jump straight to token activity.

Best hidden feature: the Codex status line can show rate limits, context stats and token counters all the time, so you never have to ask. Docs

3. Gemini CLI: /stats model

/stats defaults to session stats (duration, tool calls, performance). /stats model is the one you want: token counts and quota information per model. Docs

Dashboards and editor settings

4. Cursor: editor settings and the Usage page

On current plans, Cursor's included usage is one pool that refreshes weekly. Your remaining usage and next reset date show up in the editor settings and on the Usage page of your account dashboard. Worth knowing: different models draw from that pool at different rates. Docs

5. GitHub Copilot: right inside your IDE

In VS Code, Copilot shows what's included in your plan and your progress toward any limits, without opening a browser. If you're still on legacy premium-request billing, those counters reset on the 1st of each month at 00:00 UTC. Docs

Tools that read your local logs

6. ccusage

npx ccusage@latest and you're done. It reads local usage data from coding agent CLIs and turns it into daily, weekly, monthly and session reports. It covers Claude Code and Codex, plus OpenCode, Amp, Droid, Cline and more (ccusage claude daily, ccusage codex daily). Great when you want history, not just "how much is left right now." GitHub

Menu bar apps (always visible)

7. CodexBar

Peter Steinberger's open source (MIT) menu bar app for macOS 14+. It shows each provider's session, weekly and monthly windows with countdowns to the next reset, one status item per provider or merged into one icon. It reuses your existing provider sessions (OAuth, CLIs, API keys, browser cookies) instead of storing passwords. Install with brew install --cask codexbar. GitHub

8. TokenBar

TokenBar is built on CodexBar and keeps the same idea: usage and reset times for Claude, Codex, Cursor, Gemini and Copilot in one menu bar dropdown. On top of that it adds a pace check against the weekly window, a 30-day usage history, a system load panel, and a Pixel Agents panel that shows which coding agents are working, finished or waiting on you.

It's $5/month for Basic or $15 lifetime for Pro, macOS 14+. It reads your limits; it doesn't give you extra quota. tokenbar.site

Quick picker

The pattern I'd actually use

A terminal command for "can I start this big task?", plus one always-visible indicator so the limit never surprises you mid-run. The built-ins answer the first question well. The second is where a menu bar app earns its spot.

What's your setup? Do you check limits before a long agent run, or just find out when it stops?